## Making WAR Work for MVPs

Not everyone like Wins Above Replacement as a statistic. It is not without its flaws, like anything. Consider me among the crowd who believe it, warts and all, comes as advertised. Russell Carlton of Baseball Prospectus did a fine job breaking down their in-house Wins stat (WARP) today for those not mathematically inclined.

One challenge the more narrative-minded among us have with composite stats like WAR and WARP is the concept of all hits being created equal. That a hit is a hit is a hit, no matter when it occurs. Players cannot receive credit for driving in runners they did not place into scoring position in the first place is a stumbling block for many fans and writers. Somebody needs to come up in those spots and drive those runners home. Scoring runs, after all, is the point of the game.

Enter Win Probability Added. WPA doesn’t care how or how many. WPA simply tells us who increased their team’s chances of winning the most. The context of the situation does matter. A walkoff grand slam does more to alter the outcome of a game than a lonely solo shot in a blow out and WPA accounts for that.

What if we used WPA instead of Batting Runs to calculate WAR? Shouldn’t the ability to deliver, over and over, when the team needs it most be reflected in a player’s value?

Some warning about the slippery slope above which we stand (and where we stood before): this is not exactly mathematically sound practice. For one: WPA needs to be scaled up to equal the Runs component of WAR. More pressingly: WPA isn’t as much of a True Talent stat. It is 100% situational and, as we all know, these situations are not distributed equally. But this is Getting Blanked not Getting CERNed so screw it. Let’s go.

Using Fangraphs brand of WAR and WPA (win probability added is the same no matter where you look), subbing WPA (times 10) changes the look of the WAR leaderboard. Here’s a look at the current top ten in all of baseball.

So Mike Trout is your AL MVP while Buster Posey gets the NL award. Glad we cleared all that up. Earthshaking stuff, no doubt.

For all his September exploits, Miguel Cabrera and his poor play in high leverage situations docks him, laying the differences between him and Mike Trout all the barer. Robinson Cano takes a 2 Win hit for his (relatively) poor performances in high leverage situations.

Here are two lists, the top 15 hitters in each league by wpaWAR, National League first.

And the AL.

Torii Hunter? Welp, there goes the smell test. Admittedly, Hunter’s 2012 is very strong, but this strong? Better than everyone but his alien teammate strong? Clutch hits don’t stop, I guess we can conclude.

And finally, the players hurt the most by their performance in the clutch.

Fun to examine but ultimately meaningless. Or is it? Of course WPA isn’t entirely skill based and tells a tiny part of the story, but WAR is a descriptive stat at its heart, not predictive. As Ryan Roberts will tell you, a single 5 WAR season doesn’t mean to suggest one is a 5 Win player going forward.

It is a record of what happened over a specific baseball season. At the very least is provides a little food for thought and discussion, something solely lacking from the ongoing ad hominem attack that is woefully undercovered AL MVP race.